Assurance: Red Score, Happy Users
The dashboard is red for the wrong reason and green for the wrong reason. Read the model's inputs, live, right below.
Symptom
Assurance flags a switch red while its users are fine, and a genuinely degraded uplink shows healthy. The model is only as good as its telemetry.
Your mission
Inspect, recover, then verify
Format
Linux shell
Investigate before you configure.
Use the CLI evidence to isolate the fault, make the smallest safe correction, then verify the network state changed.
Observe the symptom and link state.
Diagnose by comparing the protocol evidence.
Verify the expected device state and confirm the original symptom is resolved.
Console access: Linux
This workspace stacks for portrait phones. Rotate to landscape for the full split-screen console.
Need a hint?
Reveal the root cause when you're ready.
Try the investigation first, then use this as your escape hatch, not a dead end.
Lab debrief
Turn the session into a repeatable troubleshooting pattern.
Use this reference after your attempt: first explain the symptom, then verify the evidence, then confirm the repair.
The problem
Assurance shows a switch's health score dropping into the red and flags a client onboarding issue, but users on that switch report no problems. Meanwhile a genuinely degraded uplink shows a healthy score.
What you'll practice
- Read AI/ML-driven health scores and their KPI breakdown in Assurance
- Distinguish a noisy signal from a real, user-impacting fault
- Verify the telemetry sources feeding a model are complete
- Drive closed-loop remediation via the Intent (REST) API
- Interpret model-driven insight critically, knowing its blind spots
The topology
A live Catalyst Center appliance orchestrates a real fabric and emits the telemetry and assurance data the AI-ops workflows analyse, so health scores are drawn from genuine, and sometimes incomplete, network state.
Commands to run yourself
The real diagnostic commands for this lab, copy-paste ready. Run these in the same order to reproduce the investigation on your own session.
curl -s -H "X-Auth-Token: $TOKEN" https://dnac.lab.example/dna/intent/api/v1/network-health | jq .curl -s -H "X-Auth-Token: $TOKEN" https://dnac.lab.example/dna/intent/api/v1/device-health | jq '.response[] | {name:.name, score:.overallHealth, updated:.lastUpdated}'curl -s -H "X-Auth-Token: $TOKEN" https://dnac.lab.example/dna/intent/api/v1/issues | jq '.response[] | {name:.name, priority:.priority, status:.status}'Topology diagram
Fact-checked references
The commands and behaviour in this lab were checked against these primary references.
Frequently asked
Is the AI just wrong?
Not exactly. A health score is only as good as its inputs: a noisy KPI can inflate a false alarm while a fault with no telemetry stays invisible. The skill is reading the score against the raw data.
What does this have to do with certifications?
It tracks Cisco's AI-operations direction (AgenticOps and the DevNet AI-Infrastructure specialist) rather than a single legacy exam, so it stays current as the platform evolves.
Ready to run this lab yourself?
No setup, no image sourcing. Book a session or ask for a live demo.